Summary of Work: Lung and bladder cancers have strong associations with environmental exposures, making them appropriate tumors in which to test the hypothesis that different environmental exposures cause different patterns of mutations in gene that initiate carcinogenesis. We have designed a series of studies to test this hypothesis. For lung cancer this includes studies of uranium miners in Colorado and Czechoslovakia; nickel smelter workers, asbestos-exposed workers, and vinyl chloride- exposed workers in Norway; and smokers and non-smokers from the U.S. For bladder cancer it includes studies of arylamine-exposed workers in Georgia and of DuPont workers in New Jersey, cyclophosphamide-treated patients from Denmark, and smokers and non-smokers from the U.S. We have published a series of papers on ras, p53, and p15 andp16 mutations and LOH from these studies. Efforts in the past year have been directed at developing new clinical collaborations and the laboratory development of comparative genomic hybridization (CGH) from fixed tissue. With the Pulmonary Medicine Clinic at University of North Carolina we are developing a study of the molecular events in early preneoplastic lesions utilizing fluorescent bronchoscopy. Fluorescent bronchoscopy allows detection of dysplastic lesions not visible using regular light bronchoscopy, and such lesions can be biopsied and followed over time. In addition we are developing a collaboration with the Mayo Clinic to provide lung tumor samples from interesting subgroups: non-smokers, and patients with first degree relatives with lung cancers as well as population-based samples. On the laboratory side, Dr. Packenham has been working with selected fixed bladder cancer samples to develop degenerative oligonucleotide priming (DOP) PCR to allow us to do CGH on fixed tissue specimens.